How to use the Dialogue Generator.
Give each speaker a goal and something they are reluctant to say. Distinct voices should come from perspective and rhythm, not exaggerated accents.
Make the workflow fit your task.
Give each speaker an immediate goal and different information. Draft the exchange through requests, evasions and reactions, then remove lines that merely tell both characters facts they already know.
- What you provide
- Scene goals and character voices.
- What you get
- Dialogue advancing the scene with distinct voices.
See the input and the result.
Illustrative input and output · a teaching example, not a live WebAct run
Example input
A mechanic wants more time; a captain wants to leave before dusk.
Completed example
Captain: Can we sail by dusk? Mechanic: You can leave, or you can arrive. Give me another hour.
Load this input into the prompt, then copy it to WebAct to try the task. Your result may differ from the illustration.
Decisions and troubleshooting.
How can speakers sound different without exaggerated accents?
Vary sentence length, directness and what each person notices or avoids. Voice should reflect perspective as well as vocabulary.
Why does the conversation sound like exposition?
Move shared facts into action or conflict. Let one speaker misunderstand, resist or discover something so the information has a reason to appear.
Try it with your own source.
Replace the example with your material in the task prompt. Keep the requirements you need, then copy the task into WebAct.
Customize and copy the task ↑